socialnetworkanalysis.Rmd

---
title: "socialnetworkanalysis"
author: "Jie Chen"
date: "8/19/2020"
output: html_document
---

# Social Network Analysis

Tutorials: 
https://eehh-stanford.github.io/SNA-workshop/intro-igraph.html
https://davetang.org/muse/2017/03/16/matrix-to-adjacency-list-in-r/#:~:text=An%20adjacency%20list%20is%20simply,list%20from%20a%20correlation%20matrix.
https://davetang.org/muse/2017/03/16/matrix-to-adjacency-list-in-r/

```{r}
# Load liraries
library("dplyr")
library("igraph")
library("reshape2")

# Import data
sna_table <- read.csv("data/sna_table.csv")
rownames(sna_table) <- sna_table[,1]
sna_table <- sna_table[,-1]
sna_table_half <- sna_table
sna_table_half[lower.tri(sna_table_half, diag = TRUE)] <- NA

# Distance table using euclidean method
dst <- sna_table%>% dist(method = 'euclidean',upper = T,diag = 0) %>% as.matrix()

# Geting target and source nodes
my_adj_list <- melt(dst)

# Generate the plot
net <- graph.data.frame(my_adj_list, directed = FALSE)
#jpeg(filename='graph.jpg',width=2000,height=2000,units='px')
plot(net, layout = layout.fruchterman.reingold, vertex.size=2, vertex.label=NA, edge.color=grey(0.85), edge.width = E(net)$weight)
#dev.off()

## retain only the connected component
net_connected <- induced.subgraph(net, subcomponent(net,1))
plot(net_connected, vertex.color="lightblue", vertex.size=5, vertex.label=NA)
```